Investigating Best Capacity Scaling Policies for Different Reconfigurable Manufacturing System Scenarios

Abstract This research presents a System Dynamics approach to model and analyze a single stage Reconfigurable manufacturing system (RMS). The model is a continuous time model. The system is exposed to a random demand that is assumed to follow a normal distribution pattern. Scaling capacity up or down is assumed unrestricted, and no outsourcing is allowed. New modifications to the existing state of the art capacity scaling model are applied in order to bring it closer to reality. A cost model for evaluating different scaling policies is introduced and a module for considering seasonal demand is added. The full-fledged simulation model was developed and tested using Vensim DSS Double Precision 5.2a package. Comprehensive experimentation and analysis are applied to evaluate the performance of five capacity scaling policies under different system scenarios. The unit cost is the performance measure considered for policies assessment. Experimentations are applied on three stages; preliminary experimentation to select the effective factors among all factors, Taguchi fractional factorial design to select significant factors among the effective factors, and 2 4 full factorial design to conduct multiple system scenarios that are used in the policy assessment process. Policy selection rules are produced based on the full factorial design results to help a practitioner in deciding the best scaling policy according to the existing system scenario. The results show that chasing demand policy and inventory-based policy have the best performance in most system scenarios.